Multiple Rumor Source Recognition Model Based on Graph Attention Networks

doi:10.19678/j.issn.1000-3428.0069882

Abstract

Abstract:

The accurate recognition of rumor sources can help suppress the spread of rumors and reduce their impact on the public. Existing rumor source recognition models often overlook the differences in mutual influence between nodes, which leads to equal weighting when aggregating neighboring feature information, thereby reducing the accuracy of rumor source recognition. This paper proposes a multiple-rumor source recognition model based on Graph Attention Networks (GATs), called, MRSDGAT. First, in a social network where a rumor has already spread, user status, source prominence of rumors, and centrality are used to represent user nodes as vectors that are then used to construct a feature matrix for the nodes. Subsequently, using the GAT is employed to explore the mutual influence between nodes, calculate the influence weights, and aggregate node feature information according to the weight of the influence between nodes. Simultaneously, residual connections are introduced between the attention layers to resolve the issue of gradient disappearance and improve the ability to identify multiple rumor sources. Finally, the model outputs the probability value of each node as a source node. The larger the probability value, the greater the possibility that the node is a source node. The experimental results show that on the Karate dataset, the F1 value of the MRSDGAT model improves by 14.09, 13.32, and 13.10 percentage points compared to the baseline GCNSI model, and by 23.41, 22.59, and 24.21 percentage points compared to the baseline LPSI model, indicating better recognition performance.

Key words: node representation, rumor source, multiple rumor source recognition, Graph Attention Networks (GAT), residual connection

摘要：

谣言源的准确识别能够抑制谣言的传播扩散, 减少谣言对社会公众的影响。目前的谣言源识别模型忽略了节点之间影响力的差异性, 导致在聚合邻居特征信息时权重相同, 从而降低了谣言源识别的准确性。提出一种基于图注意力网络(GAT)的多谣言源识别模型——MRSDGAT。首先, 在一个完成谣言传播的社交网络中, 利用用户状态、谣言来源突出性和中心性将社交网络中的用户节点表示成向量, 通过此向量构建出节点的特征矩阵。然后, 通过GAT挖掘节点之间的相互影响力, 计算节点的影响力权重, 并按照节点间的影响力权重聚合节点特征信息。接着, 在注意力层之间引入残差连接以缓解梯度消失问题, 提高识别多个谣言源的能力。最后, 模型输出的值为每个节点作为源节点的概率值, 概率值越大, 表明该节点作为源节点的可能性越大。实验结果表明, 在Karate数据集上, MRSDGAT模型的F1值比基线GCNSI模型提升了14.09、13.32、13.10百分点, 比基线LPSI模型提升了23.41、22.59、24.21百分点, 识别性能更优。

关键词: 节点表示, 谣言源, 多谣言源识别, 图注意力网络, 残差连接

MA Manfu, YANG Xin, LI Yong, LIU Zezheng. Multiple Rumor Source Recognition Model Based on Graph Attention Networks[J]. Computer Engineering, 2026, 52(2): 101-109.

马满福, 杨鑫, 李勇, 刘泽政. 基于图注意力网络的多谣言源识别模型[J]. 计算机工程, 2026, 52(2): 101-109.

/ Recommend / Download Citations

URL: https://www.ecice06.com/EN/10.19678/j.issn.1000-3428.0069882

https://www.ecice06.com/EN/Y2026/V52/I2/101

Figures/Tables 15

Fig.1 SIR infection model

Fig.2 MRSDGAT model architecture

Fig.3 MRSDGAT attention layer structure

Fig.4 MRSDGAT experimental results

Fig.5 Comparison of results with different hidden layer sizes

Fig.6 Comparison of results with different number of attention heads

Fig.7 Comparison of results with different Dropouts

Fig.8 Comparison of results with different learning rates

Fig.9 Comparison of results with different number of attention levels

References 34

1	KOSTKA J, OSWALD Y A, WATTENHOFER R. Word of mouth: rumor dissemination in social networks[C]//Proceedings of SIROCCO'08. Berlin, Germany: Springer, 2008: 185-196.
2	KURKA D B, GODOY A, VON ZUBEN F J. Online social network analysis: a survey of research applications in computer science[EB/OL]. [2024-04-08]. https://arxiv.org/abs/1504.05655.
3	ZUBIAGA A , AKER A , BONTCHEVA K , et al. Detection and resolution of rumours in social media. ACM Computing Surveys, 2019, 51 (2): 1- 36.
4	SHAH D , ZAMAN T . Rumors in a network: who's the culprit?. IEEE Transactions on Information Theory, 2011, 57 (8): 5163- 5181. doi: 10.1109/TIT.2011.2158885
5	ZANG W Y , ZHANG P , ZHOU C , et al. Locating multiple sources in social networks under the SIR model: a divide-and-conquer approach. Journal of Computational Science, 2015, 10, 278- 287. doi: 10.1016/j.jocs.2015.05.002
6	ZEJNILOVI AC'G S, GOMES J, SINOPOLI B. Network observability and localization of the source of diffusion based on a subset of nodes[C]//Proceedings of the 51st Annual Allerton Conference on Communication, Control, and Computing. Washington D.C., USA: IEEE Press, 2013: 847-852.
7	ZHU K, CHEN Z, YING L. Catch'Em all: locating multiplediffusion sources in networks with partial observations[C]//Proceedings of the 31st AAAI Conference on Artificial Intelligence. Palo Alto, USA: AAAI Press, 2017: 1676-1683.
8	ZHU K, YING L. A robust information source estimator with sparse observations[C]//Proceedings of the IEEE Conference on Computer Communications. Washington D.C., USA: IEEE Press, 2014: 2211-2219.
9	WANG Z, WANG C K, PEI J S, et al. Multiple source detection without knowing the underlying propagation model[C]//Proceedings of the AAAI Conference on Artificial Intelligence. Palo Alto, USA: AAAI Press, 2017: 217-223.
10	DONG M, ZHENG B L, HUNG N Q V, et al. Multiple rumor source detection with graph convolutional networks[C]//Proceedings of the 28th ACM International Conference on Information and Knowledge Management. New York, USA: ACM Press, 2019: 569-578.
11	EASLEY D A , KLEINBERG J M . Networks, crowds, and markets-reasoning about a highly connected world. IEEE Technology and Society Magazine, 2013, 32 (3): 10- 30. doi: 10.1109/MTS.2013.2276667
12	吴杨, 吴国文, 张红, 等. 基于扩展传染病模型的谣言溯源. 计算机与现代化, 2022 (1): 113- 119.
	WU Y , WU G W , ZHANG H , et al. Rumor source detection based on extended epidemic model. Computer and Modernization, 2022 (1): 113- 119.
13	GUO Q, ZHANG C, ZHANG H S, et al. IGCN: infected graph convolutional network based source identification[C]//Proceedings of the IEEE Global Communications Conference (GLOBECOM). Washington D.C., USA: IEEE Press, 2021: 1-6.
14	KARAMCHANDANI N, FRANCESCHETTI M. Rumor source detection under probabilistic sampling[C]//Proceedings of the IEEE International Symposium on Information Theory. Washington D.C., USA: IEEE Press, 2013: 2184-2188.
15	WANG Z X , DONG W X , ZHANG W Y , et al. Rumor source detection with multiple observations. ACM SIGMETRICS Performance Evaluation Review, 2014, 42 (1): 1- 13. doi: 10.1145/2637364.2591993
16	钱榕, 李鑫, 刘晓豫, 等. 考虑禁言机制的谣言传播模型. 计算机工程, 2024, 50 (8): 372- 378. doi: 10.19678/j.issn.1000-3428.0068506
	QIAN R , LI X , LIU X Y , et al. Rumor spreading model considering prohibition mechanism. Computer Engineering, 2024, 50 (8): 372- 378. doi: 10.19678/j.issn.1000-3428.0068506
17	ALLEN L J S . Some discrete-time SI, SIR, and SIS epidemic models. Mathematical Biosciences, 1994, 124 (1): 83- 105. doi: 10.1016/0025-5564(94)90025-6
18	ANDERSON R M , MAY R M . Infectious diseases of humans: dynamics and control. New York, USA: Oxford University Press, 1991.
19	BONACICH P . Power and centrality: a family of measures. American Journal of Sociology, 1987, 92 (5): 1170- 1182. doi: 10.1086/228631
20	FIORITI V, CHINNICI M. Predicting the sources of an outbreak with a spectral technique[EB/OL]. [2024-04-08]. https://arxiv.org/abs/1211.2333.
21	PRAKASH B A, VREEKEN J, FALOUTSOS C. Spotting culprits in epidemics: how many and which ones?[C]//Proceedings of the IEEE 12th International Conference on Data Mining. Washington D.C., USA: IEEE Press, 2012: 11-20.
22	PRAKASH B A , VREEKEN J , FALOUTSOS C . Efficiently spotting the starting points of an epidemic in a large graph. Knowledge and Information Systems, 2014, 38 (1): 35- 59. doi: 10.1007/s10115-013-0671-5
23	ZHU K , YING L . Information source detection in the SIR model: a sample-path-based approach. IEEE/ACM Transactions on Networking, 2016, 24 (1): 408- 421. doi: 10.1109/TNET.2014.2364972
24	LOKHOV A Y , MÉZARD M , OHTA H , et al. Inferring the origin of an epidemic with a dynamic message-passing algorithm. Physical Review E, 2014, 90, 012801. doi: 10.1103/PhysRevE.90.012801
25	LI L , ZHOU J Y , JIANG Y W , et al. Propagation source identification of infectious diseases with graph convolutional networks. Journal of Biomedical Informatics, 2021, 116, 103720. doi: 10.1016/j.jbi.2021.103720
26	VELI AČG KOVI AC'G P, CUCURULL G, CASANOVA A, et al. Graph attention networks[EB/OL]. [2024-04-08]. https://arxiv.org/abs/1211.2333.
27	VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[C]//Proceedings of the 31st International Conference on Neural Information Processing Systems. New York, USA: ACM Press, 2017: 5998-6008.
28	BRODY S, ALON U, YAHAV E. How attentive are graph attention networks?[C]//Proceedings of the 31st International Conference on Neural Information Processing Systems. New York, USA: ACM Press, 2017: 1-10.
29	陈轶洲, 刘旭生, 孙林檀, 等. 基于图神经网络的社交网络影响力预测算法. 南京大学学报(自然科学版), 2022, 58 (3): 386- 397.
	CHEN Y Z , LIU X S , SUN L T , et al. Social influence prediction with graph neural network. Journal of Nanjing University (Natural Science), 2022, 58 (3): 386- 397.
30	SHAH D , ZAMAN T . Rumor centrality. ACM SIGMETRICS Performance Evaluation Review, 2012, 40 (1): 199- 210. doi: 10.1145/2318857.2254782
31	ZACHARY W W . An information flow model for conflict and fission in small groups. Journal of Anthropological Research, 1977, 33 (4): 452- 473. doi: 10.1086/jar.33.4.3629752
32	GLEISER P M , DANON L . Community structure in Jazz. Advances in Complex Systems, 2003, 6 (4): 565- 573. doi: 10.1142/S0219525903001067
33	GUIMERÀ R , DANON L , DÍAZ-GUILERA A , et al. Self-similar community structure in a network of human interactions. Physical Review E: Statistical, Nonlinear, and Soft Matter Physics, 2003, 68 (2): 065103.
34	KINGMA D P, BA J. Adam: a method for stochastic optimization[EB/OL]. [2024-04-08]. https://arxiv.org/abs/1412.6980.

[1]	Haoyang LI, Xiaowei HE, Bin WANG, Hao WU, Qi YOU. Cloud Computing Resource Load Prediction Based on Improved Informer [J]. Computer Engineering, 2024, 50(2): 43-50.
[2]	JIANG Xu, QIAN Xuezhong, SONG Wei. Distantly Supervised Relationship Extraction Combined with Residual BiLSTM and Sentence Bag Attention [J]. Computer Engineering, 2022, 48(10): 110-115,122.
[3]	SHI Zhantang, MA Yupeng, ZHAO Fan, MA Bo. Chinese Named Entity Recognition Based on CNN-Head Transformer Encoder [J]. Computer Engineering, 2022, 48(10): 73-80.
[4]	Yuelong LI, Dehua TANG, Guiyuan JIANG, Zhitao XIAO, Lei GENG, Fang ZHANG, Jun WU. Short Term Traffic Flow Forecasting Based on Dimension Weighted Residual LSTM [J]. Computer Engineering, 2019, 45(6): 1-5.

Please choose a citation manager

Content to export